GDP-mannose 4,6-Dehydratase is a Key Driver of MYCN-amplified Neuroblastoma Core Fucosylation and Tumorigenesis

doi:10.21203/rs.3.rs-4720665/v1

Download PDF

Article

GDP-mannose 4,6-Dehydratase is a Key Driver of MYCN-amplified Neuroblastoma Core Fucosylation and Tumorigenesis

https://doi.org/10.21203/rs.3.rs-4720665/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

MYCN-amplification is a genetic hallmark of ~ 40% of high-risk neuroblastomas (NBs). Altered glycosylation is a common feature of adult cancer progression, but little is known about how genetic signatures such as MYCN-amplification alter glycosylation profiles. Herein, matrix-assisted laser desorption/ionization mass spectrometry imaging (MALDI-MSI) revealed increased core fucosylated glycan abundance within human MYCN-amplified NB tumors. GDP-mannose 4,6-dehydratase (GMDS) is responsible for the first committed step of de novo GDP-fucose synthesis. High GMDS expression was found to be associated with poor patient survival, advanced stage disease, and MYCN-amplification in human NB tumors. Chromatin immunoprecipitation and promoter reporter assays demonstrated that N-MYC directly binds and activates the GMDS promoter in NB cells. When GMDS was blocked through either genetic or pharmacological mechanisms, NBs were found to be dependent upon de novo GDP-fucose production to sustain cell surface and secreted core fucosylated glycan abundance, as well as adherence and motility. Moreover, when de novo fucose synthesis was inhibited in vivo, tumor formation and progression were blocked. These critical findings identify de novo GDP-fucose production as a novel metabolic vulnerability that may be exploited in designing new treatment strategies for MYCN-amplified NBs.

Biological sciences/Cancer/Paediatric cancer

Biological sciences/Cancer/Cancer metabolism

Biological sciences/Cancer/Embryonal neoplasms

Biological sciences/Cancer/Tumour heterogeneity

neuroblastoma

glycosylation

fucosylation

spatial metabolomics

Neuroblastomas (NB) are devastating childhood cancers that account for 15% of all childhood cancer deaths (1). N-MYC is a key oncogenic transcription factor and driver of NB tumorigenesis (2). MYCN-amplification is a feature of nearly 40% of all high-risk NBs and is sufficient for risk stratifying tumors as high-risk (3). Children with high-risk disease have poor 5-year survival despite aggressive multimodal treatment (2, 4). These features highlight the need to better understand drivers of NB disease progression to inform the design of novel therapeutic strategies.

The genetic landscape of NB is well defined, but little attention has been devoted to the role of post-translational modifications in NB development and progression (1, 2). N-linked glycosylation is a major post-translational modification, whereby sugar complexes are added to asparagine residues (5). Complex sugar conjugates can alter protein conformation, localization, stability, and activity (5, 6). Although glycosylation patterns are different in many adult cancers, few studies have focused on glycosylation changes associated with pediatric solid tumors in situ (6, 7, 8, 9). In this study, matrix-assisted laser desorption ionization mass spectrometry imaging (MALDI-MSI) was performed on human NBs to identify increased core fucosylated glycan abundance within MYCN-amplified NBs (10).

Fucosylation is the addition of L-fucose, a rare mammalian sugar, to proteins and lipids (5, 11). Exogenous fucose can either be salvaged or synthesized de novo from GDP-mannose precursors (12). Conversion of GDP-mannose to GDP-fucose in the de novo synthesis pathway is mediated by two enzymes: GDP-mannose 4,6-dehydratase (GMDS) and GDP-L fucose synthase (GFUS encoded by TSTA3) (13). Core fucosylation is an important type of N-linked glycosylation, whereby fucose is added to the α1,6 position of the N-acetylglucosamine (GlcNAc) residue adjoining asparagine (5, 13). In the current study, high GMDS expression was associated with poor patient survival, advanced-stage disease, and MYCN-amplification. These findings led us to hypothesize that GMDS is a key mediator of NB core fucosylation and is critical for NB progression.

Human Samples. Formalin-fixed paraffin-embedded (FFPE) NBs were obtained from residual tumor samples archived at Cincinnati Children’s Hospital Medical Center (Cincinnati, OH). Deidentified specimens were transferred to the University of Kentucky with approval from the Institutional Review Board (75968). Nine patients with Stage 4 primary neuroblastomas were included in our analysis. Four of these neuroblastomas were MYCN non-amplified, and five were MYCN-amplified. All NB samples were Stage 4 primary tumors previously treated with chemotherapy.

Tissue Preparation and Enzyme Digestion. FFPE NB samples were prepared for MALDI-MSI analysis as previously described (Fig. 1) (10). Briefly, 4 µm slices were cut onto positively charged slides, dewaxed, and deparaffinized. Slides were then submerged in citraconic anhydride buffer (50 mL HPLC water, 25 µL citraconic anhydride, and 2 µL of 36.5–38% HCl) and boiled for antigen retrieval. Buffer was then exchanged with water prior to final full water submersion. Slides were dried in a vacuum desiccator prior to enzymatic digestion.

MALDI-MSI. For the enzymatic cleavage of N-linked glycans and glycogen, an HTX M5 robotic sprayer was used to dual spray 14 passes of PNGaseF prime (N-glycan cleavage from asparagine residue) and isoamylase (α1,6 branch cleavage for glycogen) per slide, at a concentration of approximately 0.1 µg/µL PNGaseF prime and 3 U isoamylase as previously described (10). A timsTOF fleX (Bruker) with a NyGU laser operating at 1000 Hz and 90% laser energy with 200 AU energy was used to obtain N-linked glycan spectra. Laser spot size was set to 50 µm, raster size was 50 µM, and 300 shots per pixel were collected and averaged per 50 µm pixel. Mass acquisition was set to 500 m/z-3500 m/z and tuned between runs. Files were analyzed and filtered using feature lists containing m/z values for ~ 150 glycans. Neuroblast-rich regions of interest (ROI) were identified by H&E staining from an adjacent slide with our team pathologists using the QuPath-0.4.3 image software (14). Annotated ROIs were imported and aligned using SCiLS™ Lab. Average total ion count for each specimen was analyzed using MetaboAnalyst 5.0 (15, 16). Significant n-linked glycans were identified by significance analysis of microarrays (SAM) with a q-value < 0.05 and log2 fold change > 1 (15).

Cell Culture. Cell line sources and validation information is provided in Supp. Table S1. Cells were maintained in DMEM or RPMI 1640 containing 1% Pen/Strep (Cytiva, Marlborough, MA) and 10% FBS (Biowest, Brandenton, FL) in a humidified 37°C/5% CO₂ environment. Cells were passaged at 75% confluency, and experiments were performed before passage ten.

Transfection. Human GMDS 4 unique 29-mer shRNA constructs in retroviral untagged vector (OriGene Technologies, Rockville, MD) or scramble shRNA were transfected into BE(2)-C cells with Lipofectamine LTX (Thermo Fisher, Waltham, MA). Cells were selected in growth medium with puromycin to generate stable GMDS-downregulated (shGMDS) and scramble control (shCON) cell lines. For MYCN silencing, 50 pmol/well of ON-TARGETplus MYCN siRNA (Dharmacon, Lafayette, CO) or non-silencing control siRNA were transfected into BE(2)-C cells using Lipofectamine RNAiMax (Thermo Fisher). Cells were harvested 72 h following transfection.

mRNA Quantification. Total RNA from cultured cells was isolated using the RNAqueous Total RNA Isolation Kit (Thermo Fisher). RNA was reverse-transcribed using the qScript cDNA Synthesis Kit (Quanta, Houston, TX). Reverse-transcribed RNA was amplified using Powerup SYBR Green PCR Master Mix (Applied Biosystems, Foster City, CA). All data are presented relative to 18S RNA. Primer sequences used are listed in Supp. Table S2.

Immunohistochemistry. For lectin-based glycan detection, slides were deparaffinized and antigen retrieval was performed using Dako low-pH Target Retrieval Solution (Carpinteria, CA) at 95°C for 20 min using a Biocare Medical Decloaking Chamber. Endogenous streptavidin-biotin was blocked using Vector Labs Streptavidin/Biotin Blocking Kit (Newark, CA), followed by endogenous peroxidase quenching (Dako) and incubation with Carbo-Free Blocking Solution (CFB; Vector Labs). Slides were incubated with biotin-conjugated Aleuria aurantia lectin (AAL) at room temperature, followed by Peroxidase Streptavidin (Vector Labs), and visualized with DAB (Dako). Specificity was confirmed by pre-incubation of AAL with L-fucose and negative staining in control mouse kidney tissue.

Immunohistochemistry Scoring. Slides were scanned at 20x brightfield at 0.5 µ/pixel using a Leica Aperio AT2 scanner. SVS files were annotated using QuPath V.0.4.3 (14). The necrotic percentage of each tumor was determined by specifying the necrotic areas in each tumor and dividing the sum of the necrotic area by the total area of the tumor (17). Histological interpretations were reviewed by team pathologists. The percentage of AAL-positive cells was determined using QuPath’s positive cell detection with a single threshold over the entirety of each tumor.

Immunoblotting. Cells were lysed in LSD buffer with complete Protease Inhibitor Cocktail (Roche, Basel, Switzerland). Lysate was then centrifuged to remove cellular debris, resolved on a 4–12% NuPAGE gel, transferred to nitrocellulose, and immunoblotted as follows. Membranes were blocked with 0.1% casein, incubated with the primary antibody or AAL overnight, washed, incubated with secondary, washed again, and then quantified using an Odyssey imaging system (LI-COR Biosciences, Lincoln, NE). Floating and adherent cells were collected to assess levels of cleaved PARP and cyclin D1. Reagents are further described in Supp. Table S3.

Lectin-based Flow Cytometry. Cells were lifted using Versene (Gibco, Billings, MT), pelleted, resuspended in ice-cold PBS, aliquoted into individual tubes, and incubated with Zombie fixable dead cell dye (BioLegend, San Diego, CA). Non-specific binding was then blocked by incubation with CFB, and biotinylated lectins (AAL and SNA) were diluted in CFB prior to incubation with cells at room temperature. An excess of CFB was then added to each tube before cells were pelleted and incubated with avidin-conjugated Cy5 diluted in flow cytometry buffer (Hanks buffered salt solution without Ca²⁺/Mg²⁺, 1% heat-inactivated FBS, 25 mM HEPES, 5 mM EDTA). Cells were then washed and briefly fixed with 4% paraformaldehyde. Each experiment was performed at least three times with three technical replicates per sample. Fluorescence was measured on a BD Symphony, and analysis was performed using FlowJo v.10 (Treestar, Woodburn, OR).

Lectin-based ELISA. To quantify core fucosylated proteins in cell supernatants, 2 × 10⁵ cells were seeded in serum-free medium in a 24-well tissue culture plate and incubated for 20 h. Floating cells were then pelleted, and the cell-free medium was frozen at − 20°C. To analyze the effect of 2-FF on the secretion of core-fucosylated proteins, BE(2)-C cells were treated in complete medium with indicated concentrations of 2-FF for 48 h, lifted using Versene, seeded in serum-free medium containing additional 2-FF, and incubated for another 20 h before freezing the supernatant. Brefeldin A (BFA; BioLegend) was added as a control for the final 20 h of culture.

Thawed supernatants were added in duplicate to each well of a high binding plate (Greiner Bio-One, Kremsmünster, Austria), and incubated overnight at 4°C. Serum-free medium was used as a negative control. Plates were washed using PBS/0.05% Tween (PBS-T) and blocked with CFB/0.05% Tween. AAL diluted in CFB was added to each well and incubated for 1 h. Plates were washed and incubated with avidin-HRP diluted 1:2000 in PBS before washing and development with TMB.

Dual Luciferase Reporter Assay. Human genomic DNA was extracted from BE(2)-C cells, and the GMDS gene promoter was amplified from the extracted genomic DNA using a Taq Red Master Mix kit (APEX). The PCR product was cloned 5’-upstream of a luciferase reporter gene into pGL4.10[luc2] (Promega, Madison, WI) by restriction enzyme digestion using NheI, HindIII, and T4 ligase (New England BioLabs, Ipswich, MA). The constructed promoter plasmid was sequenced by ACGT. SH-EP MYCN-inducible cells were transfected with the GMDS promoter reporter and pGL4.73[hRluc/SV40] vector. After 24 h, MYCN was induced with doxycycline (100 ng/mL) for 48 h. Cell lysates were used to measure relative luciferase activity using the Dual Luciferase Reporter Assay System (Promega), and Renilla luciferase was used as a control.

ChIP Assay. Chromatin immunoprecipitation (ChIP) was performed using a Simple ChIP kit (Cell Signaling Technology, Danvers, MA). In brief, BE(2)-C cells were fixed with 1% formaldehyde; after cross-linking DNA to protein, nuclei preparation, and chromatin digestion, 10 µg of chromatin was incubated with 2 µg anti-N-MYC (Santa Cruz Biotechnology, Dallas, TX), histone H3 antibody (Cell Signaling Technology) or normal rabbit IgG (Cell Signaling Technology) at 4°C overnight. Protein G beads were used for immunoprecipitation. The DNA was purified by immunoprecipitation and quantified by qPCR. APEX promoter was used as a positive control for N-MYC pulldown (18).

Viability Determination. Viability was determined using the Cell Titer Glo Luminescent Viability Assay (Promega). in accordance with manufacturer’s instructions.

Adhesion Assay. Vybrant cell adhesion assays were performed according to the manufacturer’s instructions (Invitrogen) using a monolayer of BE(2)-C shCON cells.

Wound Healing Assay. BE(2)-C cells were seeded in each chamber of an ibidi 2-well 35 mm µ-Dish in complete medium and allowed to adhere overnight. The silicone insert was then removed and serial brightfield images were obtained using an Evos FL (Life Technologies, Carlsbad, CA) at 4×. Images were analyzed using ImageJ with the wound healing size tool, and the percentage of each gap closed was compared (19).

In vitro Treatments. Cells were incubated with either 2-FF (SynChem; Elk Grove Village, IL) or equivalent DMSO control. L-fucose was purchased from Vector Laboratories. For rescue trials, cells were incubated for three days in normal medium supplemented with L-fucose (50 and 150 µM).

For pharmacological MYC downregulation, BE(2)-C Cells were seeded and allowed to adhere overnight. The following day, medium was replaced with fresh growth medium containing either 100 nM JQ1 (Selleck Chemicals, Houston, TX) or 0.01% DMSO. Cells were incubated for 5 days before the total cell lysate was harvested for Western blotting.

Subcutaneous Xenograft Trials. Athymic nude mice (Hsd:Athymic Nude-Foxn1^nu) were purchased from Inotiv (West Lafayette, IN) and maintained at a barrier facility at the University of Kentucky. Tumors were initiated at 6–8 weeks of age, and mice were humanely euthanized when tumor diameter exceeded 1.5 cm in the greatest dimension. All procedures were approved by the University of Kentucky IACUC (2021–3934).

For our genetic knockdown model, 1 × 10⁵ shCON or shGMDS cells were resuspended in Matrigel (Corning, Corning, NY) and injected into the flank of male athymic nude male mice (n = 5 per group) using a 27-gauge needle. Mice were monitored daily. Xenograft volume was estimated using the following formula [(length × width²)/2] (20). Serial tumor volume measurements were compared using a mixed-effects model in GraphPad Prism v.10 using fixed effects of treatment and day with a random effect of mouse.

For 2-FF trials, 1 × 10⁵ BE(2)-C cells were implanted with Matrigel into athymic female mice, and tumors were allowed to reach a volume of 75 mm³. Tumor-bearing mice were then randomized to receive 2-FF (50 mM)-supplemented water or vehicle control (n = 5 mice each) (21) and monitored daily to assess tumor volume, water intake, activity, and body mass.

Statistical Analysis. Graph Pad Prism was used to perform statistical analyses, and the tests are indicated in each figure legend. In all cases, p-values < 0.05 were considered significant and are indicated in the figures by asterisks: *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001.

MALDI-MSI Reveals Increased Abundance of Complex and Core Fucosylated N-linked Glycans in Human NBs. The N-linked glycome of MYCN-amplified NB tumors was profiled using MALDI-MSI to compare MYCN-amplified and MYCN non-amplified Stage 4 human NB primary tumors. Five MYCN-amplified and four MYCN non-amplified human NB tumors were evaluated (Fig. 1). As anticipated, marked heterogeneity was noted in the histopathological analysis of human NB tumor tissues. Neuroblast-containing regions were identified using slide-adjacent H&E sections and overlaid onto the MALDI-MSI spectra (Fig. 2A). MALDI-MSI identified 59 N-linked glycans in human NB samples, based on internal standards. A heatmap of the top 25 differentially expressed N-linked glycans between MYCN-amplified and MYCN non-amplified NBs is shown in Fig. 2B. Eight N-linked glycan species showed a two-fold increase in abundance in MYCN-amplified NBs (Fig. 2C). Significance analysis of microarrays revealed 20 N-linked glycans that were significantly different between our two cohorts, featuring a q-value < 0.05 (Fig. 2D). Overall, seven differentially abundant N-linked glycans met our fold change and statistical thresholds. All seven differentially expressed glycans were enriched in MYCN-amplified NBs. Six of these glycans feature core fucosylation and five are complex in nature (Table 1). MYCN-amplified tumors also showed an increased abundance of two sialylated N-linked glycan species. Overall, MALDI-MSI revealed an increase in a subset of core fucosylated glycans within the neuroblast-rich regions of human MYCN-amplified NB specimens.

High De Novo Fucose Synthetic Gene Expression is Associated with Advanced Stage NB and Poor Overall Survival. Identifying increased core fucosylated glycan species within MYCN-amplified NBs prompted exploration of the Kocak and SEQC databases within the R2 microarray analysis and visualization platform (http://r2.amc.nl) (22, 23). This study aimed to determine whether the expression levels of key drivers of de novo GDP-fucose synthesis genes (GMDS, TSTA3), GDP-fucose transport (SLC35C1), and fucose salvage pathway (FUK, FPGT) are associated with poor overall patient survival in human NB tumors (22, 24). Kaplan–Meier analysis demonstrated that high expression of both de novo GDP-fucose synthesis genes (GMDS and TSTA3) was associated with poor overall patient survival (Bonf p < 0.0001, Bonf p < 0.0001; Fig. 3A,D; Supp. Fig. S1A). High levels of GMDS and TSTA3 were also observed in Stage 4 tumors (p < 0.0001, p < 0.0001; Fig. 3B,E; Supp. Fig. S1B), and MYCN-amplified NBs (p < 0.001; Fig. 3C,F; Supp. Fig. S1C). The expression level of SLC35C1, the GDP-fucose transporter in the Golgi, was not associated with poor patient survival. FPGT encodes a key fucose salvage pathway enzyme and low FPGT expression was associated with poor patient survival (Bonf p < 0.001; Supp. Fig. S1D). Lower FPGT levels were also observed in Stage 4 NBs (p < 0.001; Supp. Fig. S1E) and within MYCN-amplified NBs (p < 0.0001; Supp. Fig. S1F). In summary, high expression of de novo GDP-fucose synthetic genes is associated with advanced stage disease, MYCN-amplification, and poor overall survival in independent and robust human NB cohorts. GMDS was thus selected as a potential metabolic vulnerability for further investigation, given its association with severe NB clinical features and its role as the first committed step of de novo GDP-fucose synthesis.

GMDS is a Direct Transcriptional Target of N-MYC. Identifying that MYCN-amplified NBs have higher GMDS expression and a higher abundance of core fucosylated glycans led us to hypothesize that N-MYC may alter core fucosylation by directly regulating GMDS expression. Modulation of MYCN using MYCN tet-on and tet-off inducible SHEP cells showed a direct correlation between MYCN expression and both GMDS mRNA and protein expression (Fig. 4A,B). The activation of the GMDS promoter by high MYCN expression was assessed using a luciferase promoter assay. In brief, the GMDS promoter was adjoined to a luciferase reporter and transfected into MYCN tet-on SHEP cells, where induction of MYCN increased GMDS promoter activity 6-fold (Fig. 4C). The GMDS promoter features 11 non-canonical E-boxes as candidate binding sites for N-MYC activation. To determine whether N-MYC directly interacts with the GMDS promoter, we conducted a ChIP assay targeting N-MYC in MYCN-amplified BE(2)-C cells. This assay revealed enrichment of the GMDS promoter sequence within the DNA fragments pulled down by N-MYC, supporting the hypothesis that there is a direct interaction between N-MYC and the GMDS promoter. (Fig. 4D). APEX, a known N-MYC target gene, was used as the positive control. Together, these results support a direct mechanism through which N-MYC interacts with and activates the GMDS promoter.

Transient MYCN knockdown and chemical blockade using bromodomain inhibition was used to evaluate whether oncogenic N-MYC signaling alters core fucosylated abundance in NBs. Transient knockdown of MYCN using siRNA in our MYCN-amplified BE(2)-C cell line resulted in significant reduction in GMDS mRNA, as well as a 75% reduction in core fucosylation (Fig. 4E,F). There was no concurrent change in sialylation, suggesting specificity of the N-MYC/GMDS interaction in modulating core fucosylation. Bromodomain inhibition via JQ1 similarly blocked N-MYC expression and GMDS expression in association with a 50% reduction of AAL fluorescence (25, 26) (Supp. Fig. S2). These findings demonstrate that N-MYC is a direct activator of the GMDS promoter and provides a rationale for high GMDS expression and core fucosylated glycan abundance within human MYCN-amplified NB tumors.

GMDS is a Key Mediator of NB Core Fucosylation and Promotes NB Migration, Adhesion, and Tumor Formation. Genetic knockdown of GMDS (shGMDS) was performed in MYCN-amplified BE(2)-C cells to evaluate whether de novo GDP-fucose synthesis was a key mediator of NB core fucosylation and survival (Fig. 5A). Lectin-based Western blotting, flow cytometry, and ELISA revealed that GMDS-knockout cells had lower whole cell, cell surface, and secreted core fucosylated glycans (Fig. 5B–D). Exogenous L-fucose media supplementation (50 µM and 150 µM) rescued the cell surface and secreted core fucosylated glycan abundance (Fig. 5E), demonstrating the supplemental use of the fucose salvage pathway in these cells.

Knockdown of GMDS also altered NB cell attachment and migration. A two-fold decrease in cellular adhesion was observed in shGMDS cells using the Vybrant™ adhesion assay (Fig. 5F). When migration was assessed via wound healing assay, shGMDS cells moved significantly slower than control cells, failing to close a uniform gap after 64 h (Fig. 5G; Supp Fig S3). Overall, shGMDS NB cells feature depleted core fucosylated glycans and show defects in migration and adhesion in vitro.

The role of GMDS in MYCN-amplified NB tumor formation was assessed by measuring subcutaneous tumor formation following shGMDS and shCON implantation in the flanks of athymic nude mice. Mixed-effects modeling revealed that the shGMDS tumor volume was significantly lower than that of control tumors (p = 0.009; Fig. 5H). Tumor explant masses were also decreased by three-fold within the shGMDS group (p < 0.05; Fig. 5I). Three out of five shGMDS inoculations formed tumors in the trial, whereas all shCON injections yielded tumors (p = 0.44). H&E staining revealed no gross histological differences between the tumors. Together, these data suggest that NBs are reliant upon GMDS and de novo GDP- fucose synthesis for core fucosylated glycan production, tumor growth, and migration. Therefore, GMDS is a key contributor to MYCN-amplified NB tumor development, inviting further investigation into the efficacy of chemically blocking fucosylation in preclinical models of NB.

2-Fluorofucose Blocks NB Core Fucosylation and Abrogates Tumor Growth In Vivo. 2-FF is a fucose analog that is passively taken up by the salvage pathway of GDP-fucose production and inhibits both de novo fucose biosynthesis as well as fucosyltransferase activity (21, 27). Thus, 2-FF was used as a chemical tool to evaluate whether the small-molecule inhibition of fucosylation impacted NB viability and growth. This compound successfully blocked NB core fucosylation, as detected via Western blotting, flow cytometry, and ELISA, demonstrating global depletion of core fucosylated glycans (Fig. 6A–C). Flow cytometric evaluation showed a clear dose-dependent effect, with 150 µM treatment resulting in a loss of as much as 90% of AAL binding intensity in live cells. Similar to GMDS genetic knockdown, 2-FF addition to BE(2)-C cells elicited a 4-fold decrease in cellular adhesion compared to vehicle-treated controls (Fig. 6D). Five-day 2-FF incubation inhibited NB cell viability in MYCN-amplified BE(2)-C, SK-N-DZ, and CHP-212 cells (Supp. Fig. S4A). Treatment of BE(2)-C cells induced cell death in vitro as evidenced by the induction of PARP cleavage at day 3, as well as a loss of Cyclin D1 at the same time point (Supp. Fig. S4B). In summary, 2-FF is a fucosylation inhibitor that induces viability and adhesive defects similar to GMDS genetic knockdown and is an appropriate chemical probe for investigating whether fucosylation blockade is an effective strategy for blocking in vivo NB tumor growth.

2-FF has previously been used to investigate the impact of chemical de novo fucose synthesis blockade on tumor progression in established tumor models (21, 28). To evaluate the effect of the pharmacological blockade of core fucosylation on NB tumor progression, MYCN-amplified tumors were established to 75 mm³ and then randomized to 2-FF supplemented (50 mM) water or vehicle control. 2-FF water supplementation decreased tumor volume by 2.5-fold compared to vehicle-treated control mice (p < 0.05; Fig. 6E). Kaplan–Meier analysis also revealed prolonged survival with 2-FF treatment using a humane endpoint of maximal tumor dimension > 1.5 cm (p < 0.05; Fig. 6F). Treatment mice demonstrated no changes in behavior or adverse effects. Similar water consumption and body weight were observed in both groups throughout the trial. AAL lectin-based immunohistochemistry demonstrated that 2-FF treatment decreased the abundance of core fucosylated glycans by nearly 75%, and histological evaluation demonstrated large areas of treatment-effect necrosis encompassing over half of the examined tumor area, compared to less than 20% of total tumor area in vehicle controls (Fig. 6G–I). Although survival analysis of 2-FF-treated mice was terminated at day 18 due to compound expense, these data nonetheless show that 2-FF was well-tolerated and impeded MYCN-amplified tumor growth via the induction of tumor cell death within our established NB tumor model.

Historically, glycome profiling of solid tumors has largely been performed on cell lines in vitro or using whole tissue homogenates that may obscure or mask differential glycan abundance within heterogeneous solid tumors, such as neuroblastomas. However, recent advances in imaging mass spectrometry have allowed tissue heterogeneity to be preserved, thus permitting comparison of unique histologically-comparable regions of interest across patient samples (16, 29, 30). Herein, we performed MALDI-MSI to profile N-linked glycans in situ within human NBs. Given the master regulatory role of oncogenic MYC signaling, we sought to define the N-linked glycome differences between MYCN-amplified and non-amplified tumors and focused our initial characterization on the analysis of neuroblast-rich tumor regions that feature this genetic signature (31). Our analysis revealed an increased abundance of core fucosylated glycans as a unique feature of the human MYCN-amplified NB glycome. Initial identification of unique metabolic end products within human MYCN-amplified tumors informed exploration of robust existing and publicly shared RNA sequencing repositories to guide selection of de novo fucose synthesis as a novel therapeutic target for further investigation (22, 24).

Although increased core fucosylation is a common feature of many advanced adult cancers, the role of aberrant fucosylation in NB development and progression has not been previously described (13, 32). This is the first in situ characterization of the N-linked glycome of human NBs focusing on MYCN-amplification, a key genetic feature critical for NB risk stratification. Oncogenic MYC contributes to malignant transformation by driving glucose, glutamine, serine, essential amino acids, and fatty acid metabolism in cancer cells (33). A guiding principle for our study design was that the metabolic transforming consequences of MYCN-amplification would elicit a unique N-linked glycan signature. Oncogenic MYC signaling has recently been reported as a driver of sialoglycan production (34). Intriguingly, core fucosylated species were the more abundant N-linked glycans identified within our MYCN-amplified NB cohort. Given the described role of oncogenic MYC in regulating both glycolytic and glutamine-derived precursors of the hexosamine biosynthetic pathway, further investigation into how oncogenic MYC levels may alter glycan precursor pools to facilitate unique glycans profiles merits further study (35).

GMDS is the first of two enzymes necessary for the conversion of GDP-mannose to GDP-fucose in the de novo GDP-fucose biosynthetic pathway. N-MYC preferentially binds to E-boxes, and the GMDS promoter features 11 non-canonical E-boxes (36). Induction and repression of N-MYC in NB cells directly correlated with GMDS expression, and promoter reporter and chromatin immunoprecipitation assays demonstrated that N-MYC has a direct role in activating the GMDS promoter. These findings provide a mechanistic link by which this high-risk NB genetic profile augments the critical mediators of de novo GDP-fucose synthesis. GMDS has previously been shown to be regulated by oncogenic C-MYC/Max signaling in Burkitt’s lymphoma (37). Herein, we extend this link to aberrant N-MYC signaling and demonstrate in vitro that RNA interference and chemical inhibitors of MYC signaling can alter the abundance of core fucosylated glycans in MYCN-amplified NBs. N-MYC dysregulation is a key feature of subsets of several other pediatric (rhabdomyosarcoma, medulloblastoma, Wilms tumor, and retinoblastoma) and adult cancers (prostate cancer, small cell lung cancer, basal cell carcinomas, breast cancer, glioblastoma multiforme) (38). Identifying whether increased core fucosylated abundance is a consistent feature of N-MYC overexpression and MYCN-amplified cancers merits further investigation and may provide a rationale for the design of clinical trials to investigate the efficacy of drugs targeting de novo GDP-fucose production.

Recognition of aberrant glycosylation as a potential mediator of cancer progression has prompted considerable efforts to create small-molecule chemical tools to allow therapeutic manipulation of the glycome (39). 2-FF is a commercially available fucose analog that is taken up through the salvage pathway but blocks fucosylation through the de novo pathway and fucosyltransferases (21). Peroral bioavailability and tolerance in animal models make it an appealing drug candidate. However, these compounds have low potency, and a phase I adult clinical trial was stopped due to a high risk of thromboembolic events (40). Several other investigative groups have recently reported new fucose derivatives that potently inhibit de novo fucose synthesis and/or fucosyltransferases, including fucostatin, Fucotrim I, and 6-alkynyl fucose (41, 42, 43). Future investigations will seek to evaluate the utility of these fucosylation inhibitors in advanced preclinical NB models.

A key limitation of our study was the limited number of clinical samples and access to clinical features in our MALDI-MSI analysis of human NBs. Correlating GMDS expression with NB patient survival, stage of disease, and MYCN-amplification within large, independent human NB cohorts strengthens our confidence that de novo fucose synthesis is critical for human NB tumor progression. However, validating our MALDI-MSI observations within larger cohorts correlated with patient outcomes is an essential next step in exploring a potential new field of pediatric solid tumor biology. FFPE tissue microarrays are a rare resource for studying pediatric solid tumors but are well-suited for interrogation with MALDI-MSI (16). Another limitation of our bench investigations was the use of an immunodeficient animal model to assess the impact of de novo fucose synthesis blockade on NB progression. We demonstrated that NB cells have decreased adhesion in our in vitro assays but have not elucidated the implications of these cell surface changes within advanced in vivo models of NB featuring competent cancer-stroma and cancer–immune cell interactions.

Acknowledgements

This research was generously supported by the University of Kentucky Center for Cancer and Metabolism, funded through the NIH/NIGMS COBRE (P20 GM121327), the American Pediatric Surgical Association Grosfeld Scholarship, and the Dick Vitale Pediatric Cancer Research Fund (V2023-026) to E.J.R.; R01CA266004 and R01 CA288696 to R.C.S and M.S.G. Technical support was provided by the Flow Cytometry and Immune Monitoring Core Facility and the Biospecimen Procurement and Translational Pathology (BPTP). Image scanning was provided by the University of Kentucky Alzheimer’s Disease Research Center (P30 AG072946). Tetracycline-inducible SHEP cells were generously provided by Dr. Rani George. We are grateful to Dana Napier for her specific efforts in performing immunohistochemistry. We appreciate Donna Gilbreath with the Markey Cancer Center Research Communications Office for their assistance with manuscript preparation.

Author Contributions

BZ, MGP, and EJR were responsible for experimental design, research conduct, data acquisition, analysis of data, data interpretation, and manuscript preparation and review. BME contributed to experimental design, data interpretation, and manuscript review. RCS and MSG led design and conduct of MALDI-MSI data acquisition and contributed with data interpretation, and manuscript preparation and review. DBA assisted with experimental design, data interpretation, pathology review, and manuscript review. JL and CW contributed to experimental design and data analysis. OLN and NRS conducted pathologic review of human and animal tissue analysis. JG assisted with experimental design. SK performed MALDI-MSI data acquisition and analysis. CS, RR, MDB, and LTB conducted MALDI-MSI data analysis and assisted with data interpretation.

Competing Interests

The authors declare no potential conflicts of interest.

Data Availability

Animal and in vitro experiments were conducted at the University of Kentucky. MALDI-MSI raw data were obtained from the University of Florida. The data analyzed in Fig. 2 and Supp. Fig. S2 were obtained from the R2 Genomics Analysis and Visualization Platform (https://hgserver1.amc.nl/cgi-bin/r2/main.cgi?open_page). The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Brodeur GM. Neuroblastoma: biological insights into a clinical enigma. Nat Rev Cancer. 2003;3(3):203-16.
Cheung NK, Dyer MA. Neuroblastoma: developmental biology, cancer genomics and immunotherapy. Nat Rev Cancer. 2013;13(6):397-411.
Park JR, Bagatell R, London WB, Maris JM, Cohn SL, Mattay KK, et al. Children's Oncology Group's 2013 blueprint for research: neuroblastoma. Pediatr Blood Cancer. 2013;60(6):985-93.
Mahapatra S, Challagundla KB. Neuroblastoma. StatPearls. Treasure Island, FL: StatPearls Publishing; 2023.
Varki A, Cummings RD, Esko JD, Stanley P, Hart GW, Aebi M, et al., editors. Essentials of Glycobiology. New York: Cold Spring Harbor 2022.
Pinho SS, Reis CA. Glycosylation in cancer: mechanisms and clinical implications. Nat Rev Cancer. 2015;15(9):540-55.
Hu Y, Mayampurath A, Khan S, Cohen JK, Mechref Y, Volchenboum SL. N-linked glycan profiling in neuroblastoma cell lines. J Proteome Res. 2015;14(5):2074-81.
Ho WL, Hsu WM, Huang MC, Kadomatsu K, Nakagawara A. Protein glycosylation in cancers and its potential therapeutic applications in neuroblastoma. J Hematol Oncol. 2016;9(1):100.
Sumer-Bayraktar Z, Fife CM, Byrne FL, Kavallaris M, Packer NH. Membrane glycome is impacted by the cell culturing mode of neuroblastoma cells with differing migration and invasion potential. Glycobiology. 2022;32(7):588-99.
Stanback AE, Conroy LR, Young LEA, Hawkinson TR, Markussen KH, Clarke HA, et al. Regional N-glycan and lipid analysis from tissues using MALDI-mass spectrometry imaging. STAR Protoc. 2021;2(1):100304.
Shan M, Yang D, Dou H, Zhang L. Fucosylation in cancer biology and its clinical applications. Prog Mol Biol Transl Sci. 2019;162:93-119.
Schneider M, Al-Shareffi E, Haltiwanger RS. Biological functions of fucose in mammals. Glycobiology. 2017;27(7):601-18.
Li J, Hsu HC, Mountz JD, Allen JG. Unmasking Fucosylation: from Cell Adhesion to Immune System Regulation and Diseases. Cell Chem Biol. 2018;25(5):499-512.
Bankhead P, Loughrey MB, Fernández JA, Dombrowski Y, McArt DG, Dunne PD, et al. QuPath: Open source software for digital pathology image analysis. Sci Rep. 2017;7(1):16878.
Xia J, Psychogios N, Young N, Wishart DS. MetaboAnalyst: a web server for metabolomic data analysis and interpretation. Nucleic Acids Res. 2009;37(Web Server issue):W652-60.
Conroy LR, Stanback AE, Young LEA, Clarke HA, Austin GL, Liu J, et al. In Situ Analysis of N-Linked Glycans as Potential Biomarkers of Clinical Course in Human Prostate Cancer. Mol Cancer Res. 2021;19(10):1727-38.
Walker RA. Quantification of immunohistochemistry--issues concerning methods, utility and semiquantitative assessment I. Histopathology. 2006;49(4):406-10.
Perini G, Diolaiti D, Porro A, Della Valle G. In vivo transcriptional regulation of N-Myc target genes is controlled by E-box methylation. Proc Natl Acad Sci U S A. 2005;102(34):12117-22.
Suarez-Arnedo A, Torres Figueroa F, Clavijo C, Arbeláez P, Cruz JC, Muñoz-Camargo C. An image J plugin for the high throughput image analysis of in vitro scratch wound healing assays. PLOS ONE. 2020;15(7):e0232565.
Rellinger EJ, Padmanabhan C, Qiao J, Craig BT, An H, Zhu J, et al. Isoxazole compound ML327 blocks MYC expression and tumor formation in neuroblastoma. Oncotarget. 2017;8(53):91040-51.
Okeley NM, Alley SC, Anderson ME, Boursalian TE, Burke PJ, Emmerton KM, et al. Development of orally active inhibitors of protein and cellular fucosylation. Proc Natl Acad Sci U S A. 2013;110(14):5404-9.
Kocak H, Ackermann S, Hero B, Kahlert Y, Oberthuer A, Juraeva D, et al. Hox-C9 activates the intrinsic pathway of apoptosis and is associated with spontaneous regression in neuroblastoma. Cell Death Dis. 2013;4(4):e586.
Zhang W, Yu Y, Hertwig F, Thierry-Mieg J, Zhang W, Thierry-Mieg D, et al. Comparison of RNA-seq and microarray-based models for clinical endpoint prediction. Genome Biol. 2015;16(1):133.
Su Z, Fang H, Hong H, Shi L, Zhang W, Zhang W, et al. An investigation of biomarkers derived from legacy microarray data for their utility in the RNA-seq era. Genome Biol. 2014;15(12):523.
Debray H, Montreuil J. Aleuria aurantia agglutinin. A new isolation procedure and further study of its specificity towards various glycopeptides and oligosaccharides. Carbohydr Res. 1989;185(1):15-26.
Delmore JE, Issa GC, Lemieux ME, Rahl PB, Shi J, Jacobs HM, et al. BET bromodomain inhibition as a therapeutic strategy to target c-Myc. Cell. 2011;146(6):904-17.
Rillahan CD, Antonopoulos A, Lefort CT, Sonon R, Azadi P, Ley K, et al. Global metabolic inhibitors of sialyl- and fucosyltransferases remodel the glycome. Nat Chem Biol. 2012;8(7):661-8.
Disis ML, Corulli LR, Gad EA, Koehnlein MR, Cecil DL, Senter PD, et al. Therapeutic and Prophylactic Antitumor Activity of an Oral Inhibitor of Fucosylation in Spontaneous Mammary Cancers. Mol Cancer Ther. 2020;19(5):1102-9.
West CA, Wang M, Herrera H, Liang H, Black A, Angel PM, et al. N-Linked Glycan Branching and Fucosylation Are Increased Directly in Hcc Tissue As Determined through in Situ Glycan Imaging. J Proteome Res. 2018;17(10):3454-62.
Butler W, McDowell C, Yang Q, He Y, Zhao Y, Hauck JS, et al. Rewiring of the N-Glycome with prostate cancer progression and therapy resistance. NPJ Precis Oncol. 2023;7(1):22.
Schwab M, Alitalo K, Klempnauer KH, Varmus HE, Bishop JM, Gilbert F, et al. Amplified DNA with limited homology to myc cellular oncogene is shared by human neuroblastoma cell lines and a neuroblastoma tumour. Nature. 1983;305(5931):245-8.
Keeley TS, Yang S, Lau E. The Diverse Contributions of Fucose Linkages in Cancer. Cancers (Basel). 2019;11(9).
Dong Y, Tu R, Liu H, Qing G. Regulation of cancer cell metabolism: oncogenic MYC in the driver's seat. Signal Transduct Target Ther. 2020;5(1):124.
Smith BAH, Deutzmann A, Correa KM, Delaveris CS, Dhanasekaran R, Dove CG, et al. MYC-driven synthesis of Siglec ligands is a glycoimmune checkpoint. Proc Natl Acad Sci U S A. 2023;120(11):e2215376120.
Stine ZE, Walton ZE, Altman BJ, Hsieh AL, Dang CV. MYC, Metabolism, and Cancer. Cancer Discov. 2015;5(10):1024-39.
Zeid R, Lawlor MA, Poon E, Reyes JM, Fulciniti M, Lopez MA, et al. Enhancer invasion shapes MYCN-dependent transcriptional amplification in neuroblastoma. Nat Genet. 2018;50(4):515-23.
Li Z, Van Calcar S, Qu C, Cavenee WK, Zhang MQ, Ren B. A global transcriptional regulatory role for c-Myc in Burkitt's lymphoma cells. Proc Natl Acad Sci U S A. 2003;100(14):8164-9.
Liu R, Shi P, Wang Z, Yuan C, Cui H. Molecular Mechanisms of MYCN Dysregulation in Cancers. Front Oncol. 2020;10:625332.
Mereiter S, Balmaña M, Campos D, Gomes J, Reis CA. Glycosylation in the Era of Cancer-Targeted Therapy: Where Are We Heading? Cancer Cell. 2019;36(1):6-16.
Do KT, Chow LQM, Reckamp K, Sanborn RE, Burris H, Robert F, et al. First-In-Human, First-In-Class, Phase I Trial of the Fucosylation Inhibitor SGN-2FF in Patients with Advanced Solid Tumors. Oncologist. 2021;26(11):925-e1918.
Pijnenborg JFA, Rossing E, Merx J, Noga MJ, Titulaer WHC, Eerden N, et al. Fluorinated rhamnosides inhibit cellular fucosylation. Nat Commun. 2021;12(1):7024.
Allen JG, Mujacic M, Frohn MJ, Pickrell AJ, Kodama P, Bagal D, et al. Facile Modulation of Antibody Fucosylation with Small Molecule Fucostatin Inhibitors and Cocrystal Structure with GDP-Mannose 4,6-Dehydratase. ACS Chem Biol. 2016;11(10):2734-43.
Kizuka Y, Nakano M, Yamaguchi Y, Nakajima K, Oka R, Sato K, et al. An Alkynyl-Fucose Halts Hepatoma Cell Migration and Invasion by Inhibiting GDP-Fucose-Synthesizing Enzyme FX, TSTA3. Cell Chem Biol. 2017;24(12):1467-78.e5.

Table 1 is available in the Supplementary Files section.

There is NO conflict of interest to disclose.

SupplementalData71024.pdf
Table1.xlsx
Table 1. Summary of significant N-linked glycans differentially expressed between MYCN-amplified and MYCN non-amplified NBs.

Download PDF

Editorial decision: revise
05 Sep, 2024
Review #2 received at journal
14 Aug, 2024
Review #1 received at journal
30 Jul, 2024
Reviewer #2 agreed at journal
24 Jul, 2024
Reviewer #1 agreed at journal
16 Jul, 2024
Reviewers invited by journal
12 Jul, 2024
Submission checks completed at journal
11 Jul, 2024
Editor assigned by journal
10 Jul, 2024
First submitted to journal
10 Jul, 2024

You are reading this latest preprint version

GDP-mannose 4,6-Dehydratase is a Key Driver of MYCN-amplified Neuroblastoma Core Fucosylation and Tumorigenesis

Status:

Version 1

Abstract

Figures

Introduction

Materials/Subjects and Methods

Results

Discussion

Declarations

References

Table

Additional Declarations

Supplementary Files

Status:

Version 1